miniblog.

← Back to all posts
Apr 28, 2016 at 18:31
Are you comparing floating point numbers? Choosing a correct value for epsilon is impossible in the general case:
Comparing Floating Point Numbers, 2012 Edition
Comparing Floating Point Numbers, 2012 Edition
This post is a more carefully thought out and peer reviewed version of a floating-point comparison article I wrote many years ago. This one gives solid advice and some surprising observations about…

Related Posts

Oct 1, 2019 at 19:05
A new 16-bit floating point format for machine learning! https://hub.packtpub.com/why-intel-is-betting-on-bfloat16-to-be-a-game-changer-for-deep-learning-training-hint-range-trumps-precision/ It increases range at the expense of precision, and often allows 16-bit computation (smaller, faster hardware) to replace 32-bit ML logic.
Why Intel is betting on BFLOAT16 to be a game changer for deep learning training? Hint: Range trumps Precision.
Why Intel is betting on BFLOAT16 to be a game changer for deep learning training? Hint: Range trumps Precision.
The paper presents a comprehensive study indicating the success of Brain Floating Point (BFLOAT16) half-precision format in Deep Learning training across image classification, speech recognition, industrial recommendation systems and more.
Oct 27, 2018 at 17:08
Interestingly, Emacs lisp considers the literal 1. to be an integer literal, whereas most languages consider a decimal point to always mean a floating point number.
Apr 1, 2016 at 07:21
It still seems odd to me that there are many different NaN values in floating point arithmetic. Those bits could have improved precision!